Sulfonylurea herbicides such as sulfometuron methyl (I) and chlorsulfuron (II) inhibit growth of some bacteria, yeast and higher plants by blocking acetolactate synthase [ALS, EC 4.1.3.18], the first common enzyme in the biosynthesis of the branched-chain amino acids valine, leucine and isoleucine. The biosynthesis of branched-chain amino acids and, hence, the toxicity of sulfonylurea herbicides is restricted to plants and microbes. ALS is also inhibited by a structurally unrelated class of herbicides, the imidazolinones. ##STR1##
Three major isozymes of ALS, designated I, II and III, have been identified in enteric bacteria, Isozymes I and III, but not II, are sensitive to end-product inhibition by valine. Each of the three bacterial isozymes comprises a large and a small protein subunit. ALS enzymes from the yeast Saccharomyces cerevisiae and from some higher plants have been partially characterized and show some degree of end-product inhibition. It is not known if the yeast and plant ALS enzymes consist of one or more different polypeptides. Evidence suggests that the cellular locations of the yeast and plant ALS enzymes are in the mitochondria and chloroplasts, respectively.
Genes encoding ALS enzymes have been isolated from the enteric bacteria Salmonella typhimurium and Escherichia coli, and the yeast S. cerevisiae. The nucleotide sequences of the genes coding for the two subunits of E. coli ALS isozymes I, II and III show that they are organized as operons ilvBN, ilvGM and ilvIH, respectively. Comparison of the deduced amino acid sequences of the large subunits of the E. coli ALS isozymes shows three regions with about 50% conserved amino acids, comprising about two-thirds of the proteins, and separated by regions sharing little discernible homology. Amino acid sequence conservation, though less extensive, is also evident among the small subunits of the bacterial isozymes. In the yeast S. cerevisiae, a single gene, ILV2, essential for ALS activity was identified. Nucleotide sequence analysis of the ILV2 gene has revealed that the polypeptide encoded by it is homologous to the large subunits of the bacterial ALS isozymes. The deduced amino acid sequence of the yeast ALS shows the same degree of structural organization and the same degree of homology as is observed between the large subunits of the bacterial isozymes, except for about ninety amino acids at the amino terminus of the yeast protein that are believed to be involved in the translocation of the protein into the mitochondrion. No information on the structure of plant genes encoding ALS or the amino acid sequence of plant ALS enzymes was available prior to the inventions disclosed herein.
Enteric bacterial isozyme I is the only ALS in nature that is known to be insensitive to inhibition by sulfometuron methyl and chlorsulfuron. Therefore, enteric bacteria are sensitive to these herbicides only in the presence of valine, which inhibits isozyme I. Sulfonylurea herbicide-resistant mutant forms of the enteric bacteria Salmonella typhimurium and E. coli (selected in the presence of valine), the yeast S. cerevisiae and the higher plants Nicotaiana tabacum (tobacco), Arabidopsis thaliana and Zea mays (corn) have been identified. These mutant phenotypes cosegregate with herbicide-resistant forms of ALS through genetic crosses. In S. typhimurium the herbicide-resistance mutations are genetically linked to a gene encoding ALS, and in E. coli and S. cerevisiae, these mutations reside in the structural genes for ALS. In the higher plants the mutations responsible for the resistance are inherited as single, dominant or semidominant nuclear traits. In tobacco, these mutations map to either of two unlinked genetic loci.
The chemical control of undesirable weeds associated with agronomically useful crops requires the use of highly selective chemical herbicides. In some cases, it is difficult to identify any chemical which kills weeds without injury to the crop plant. The introduction of herbicide-resistance as a biological trait in crop plants would overcome this difficulty.
Although many genes involved in the structure and function of differentiated plant tissues and organs are not expressed in undifferentiated tissues, those involved in basic cellular functions are expressed and can be selected for in a disorganized callus or cell suspension culture. This has been demonstrated in many cases by the selection of a phenotype in tissue culture from which plants expressing the same phenotype have been regenerated. Examples include the in vitro selection of plants resistant to herbicides, pathotoxins or diseases, antibiotics, amino acid analogues, salt tolerance, etc.
Since acetolactate synthase is an enzyme involved in the basic cellular metabolic activity of amino acid biosynthesis, it was expected and has been demonstrated that genes encoding this enzyme are expressed in callus tissue as well as the whole plant. The sulfonylurea resistant tobacco mutants described in this patent, S4, C3 and Hra, were first selected in tissue culture and subsequently regenerated into whole plants in which the resistant phenotypes were retained in a genetically stable manner. Callus tissues derived from regenerated plants or their progeny continue to grow on concentrations of the herbicide which inhibit the growth of wild type callus. Thus resistance to a sulfonylurea herbicide at the plant cellular level is predictive of resistance at the whole plant level. In addition, it has been demonstrated in bacteria, yeast and higher plants that mutations resulting in the production of herbicide resistant ALS are sufficient to confer resistance at the cellular level and, in the case of plants, at the whole plant level. Therefore, the observation of herbicide-resistant ALS in extracts of plant cells is also predictive of herbicide resistant growth of cultured plant cells and herbicide resistant growth of whole plants.
Sulfonylurea herbicide resistant mutant tobacco and corn plants have been obtained by regeneration from mutant tissue culture cell lines and resistant Arabidopsis plants have been produced by seed mutagenesis. There are, however, significant advantages to be derived from isolation of a nucleic acid fragment able to confer herbicide resistance and its subsequent introduction into crops through genetic transformation. One can obtain cross species transfer of herbicide resistance, while avoiding potential limitations of tissue culture, seed mutagenesis, and plant breeding as techniques to transfer novel DNA fragments and traits. Plants exhibiting herbicide resistance achieved through transformation with a mutant ALS gene may possess distinct advantages relative to those regenerated after selection with a herbicide in tissue culture. The insertion of an additional gene or genes encoding an altered form of the ALS enzyme in the transformed plant can supply additional plant metabolic capabilities. It can also enable the plant molecular biologist to engineer desired selectivities into the added gene(s). Further, the insertion of the additional gene(s) in particular locations can result in enhanced levels of expression of the mutant ALS enzyme, as well as in different patterns of tissue or temporal expression of the gene. Such changes may result in production of new protein in root systems, for example. Tissue specific and/or temporal expression of the introduced gene can also be modulated through the substitution of specific gene regulatory sequences for the native gene regulatory sequences. Such substitutions can, for example, place gene expression under the control of chemical inducing agents. Finally, control of the chromosomal location of the inserted gene may avoid the complications of the native gene being linked to a disadvantageous allele which would require extensive plant breeding efforts to subsequently separate the traits, and the absence of exposure of the plant tissues to mutagenic agents obviates the need for extensive back-crossing to remove undesirable mutations generated by these agents.
Although genes isolated from one plant have been introduced and expressed in other plants, nonplant genes have been expressed in plants only as chimeric genes in which the coding sequences of the non-plant genes have been fused to plant regulatory sequences required for gene expression. However, it would be difficult to introduce herbicide resistance into plants by introducing chimeric genes consisting of bacterial or yeast genes for herbicide-resistant forms of ALS, since (a) these microbial ALS enzymes are believed to lack a specific signal (transit) peptide sequence required for uptake into plant chloroplasts, the cellular location of plant ALS, (b) the bacterial isozymes consist of two different polypeptide subunits, and (c) the microbial ALS enzymes may not function optimally in the foreign cellular environment of higher plants. Therefore, there is a need for nucleic acid fragments (1) which encode a herbicide-resistant form of plant ALS, and (2) which can confer herbicide resistance when introduced into herbicide sensitive plants.